Norman's blog

Baysians vs frequentists

Sat, 06 Sep 2025 00:00:00 +0000

When I first heard that statistics had a long-running dispute between “Bayesians” and “frequentists”, I was confused: this is math; how can there be long-running disputes? In math, things are proven; the losing side loses definitively and moves on. So what gives?

It turns out the dispute is about the social role of the statistician, not the math itself. In a nutshell, if you’re the person in charge, you take a Bayesian approach, while if you’re a subordinate with a mere advisory role, frequentist is more appropriate. As an example, a frequentist might say (or might have said in the days of feudalism):

“Prithee, my lord, please kindly deign to notice that the last 37 people who tried your idea died horribly.” </blockquote>
whereas a Bayesian might say:

“Besides being obviously idiotic, this idea of yours killed the last 37 people who tried it. I don’t have time for any more of this shit; you’re fired.” </blockquote>
To explain in detail:
A common misconception about statistics relates to P values. In medicine and biology, a P value less than or equal to 0.05 is generally considered “statistically significant”, while P values greater than that are considered “statistically insignificant”. The misconception is to say that a P value of 0.05 means that “there is only a 1 in 20 chance that this result is just random noise”, 0.05 being equal to 1/20.
(There’s no magic about the number 0.05; it’s just the convention that the business has standardized on, as a compromise between the risk of ignoring real results and the risk of wrongly accepting false results. Alternative thresholds might well be better, but it’s not a ridiculous convention, and with any alternative threshold there’d also be the same misconception, just with another number rather than 20.)
I’ve seen this misconception even from some very sophisticated people, but it’s not what P values actually mean. The actual meaning is: if this were just random noise, then there’d only be a 1 in 20 chance of it happening. (Or to be really precise, a 1 in 20 chance of it or a result even more extreme happening.)
This is a much less useful and more complicated meaning. We use it because it’s what we can calculate easily and objectively – we know the math of random noise – but you have to know the context to interpret it. If, for instance, this is a “hot” research topic, and many researchers have jumped in, there will be a lot of achievements of P<0.05 even if the whole field is just measuring random noise. Monkeys typing randomly at keyboards cannot actually produce the works of Shakespeare in any realistic sense, but ten researchers measuring noise can easily produce a result with P=0.05. Of course 19 of 20 of their experiments (on average) will produce null results, but there’s a fair chance that they won’t even try to publish those, perhaps even thinking that they must have messed up their experiments rather than the new hot theory being wrong. (There are always lots of ways to mess up experiments, especially when they’re cutting-edge experiments at the frontiers of science.)
So interpreting scientific results is hard; it would be much nicer if that misconception were actually true. Bridging the gap between the misconception and the reality is what Bayes’ Theorem is for.
Frequentists don’t deny the truth of Bayes’ theorem, nor could they, since it’s exceedingly simple and easy to prove. They’d just prefer not to use it. And with good reason, since using it demands that you come up with a “prior probability”. The theorem is written as:
\[ P(A|B) = \frac{P(B|A)P(A)}{P(B)}. \]
The thing we want to know, and that the misconception says we really know, is the quantity on the left, P(A|B), “the probability of A given B”, where here we’re taking A to be “the theory is correct” and B to be “the experimental results came out the way they did.” In other words, assuming the experimental results are correct – they might not be; there might be some monkey business going on; but assuming that isn’t the case – what is the chance of the theory being correct? That’s the meaning of P(A|B). Or, well, to be precise, not “the theory is correct” – that’s way too strong a statement – but rather “the results agree with this theory”; there are always other theories that would predict the same thing in this case but different things in other cases. And usually in biology / medicine, the theory in question is the null hypothesis (“nothing here but random noise”), and we’re trying to disprove it – to make the rather modest statement that “there’s something here even if we can’t say quite what.” (The rest of the paper should address what that something might be.)
Bayes’ Theorem tells us that there are three quantities which go into computing the number that we really want to know. The first is P(B|A), which is the aforementioned “P value”. We know the math of random noise; with a few relatively modest assumptions to precisely define which sort of random noise we’re dealing with, we can easily compute that value. (Those assumptions are of the “never exactly correct but generally close enough” sort.)
The second number is P(B), the probability of B, irrespective of whether A is true. We can get it from background statistics of that particular outcome, or from a control group. So it’s also not a big problem.
But unfortunately the third quantity, P(A), is a total wild card. In Bayesian lingo, it’s the “prior probability” of our hypothesis – what we considered its chance to be, before we did the experiment. Bayes’ theorem is not just giving us an opportunity to sneak in our prejudices; it’s demanding them from us.
To illustrate, suppose someone does an experiment showing that a law of physics is wrong – and not some recent, speculative law of physics, but one of those basic laws about which Rutherford is said to have stated that you should never bet against them at odds of less than 10^{12 to one (a quadrillion to one).
He might not actually have said that, but regardless, suppose you
agree with that quote and 10^{-12 is your prior
probability of this law of physics being wrong. Then you look
into the details of the experiment, and if it were mere random
noise there’d be a one percent chance of the results coming out
that way. Then your posterior probability is 10^{-10,
even if you totally believe the experiment was done correctly and
the results reported accurately. (The number is not quite that,
since I’m ignoring P(B), but that’s the general drift.)}}}
In this case the difference between the misconceived and the real meaning of the P value is particularly stark: rather than believing there’s now only a 1% chance the law of physics is correct, you still believe there’s still something like a 99.99999999% percent chance.
Islamic extremists are also Bayesian in their attitudes: their prior probability that Islam is incorrect is zero, and thus any evidence that it might be incorrect gets multiplied by zero in their heads. However many times they run into such evidence, zero multiplied by anything is still zero.
The now-deceased owner of the yacht Bayesian was also presumably a Bayesian, as is appropriate for someone like him who was the head of a financial company. Michael Lewis, in Liar’s Poker, describes Wall Street traders calling themselves “Big Swinging Dicks”; in accordance with that theme, the world-record 72-meter mast on the boat was quite a large erection, and in a sudden squall it exerted so much leverage that it swung down to the water, capsizing the boat and killing most on board, including its owner. But while we may laugh at such people’s excesses, still, as the ones who call the shots, they should be Bayesians and use all their prior knowledge, not be shy about it.
Their subordinates, though, responsible for feeding them facts they rely on, should take a different tack. Their prejudices are not supposed to override their boss’s. They may be in training for his position, but they’re not there yet. (Well, one of them may be now.) It’s best to give him only the information they can prove, and let him make the wild extrapolations, or perhaps inject their opinions in a mild, tentative, take-it-or-leave-it type of way.
In this dispute, there’s an attempt to split the baby by using a so-called “non-informative prior”. This is a contradiction in terms; the whole point of the prior is to provide information. If you have a prior belief, use it; if I disagree with that belief, I can substitute my own and get a different result. There can be a place for a mild, compromise prior, that satisfies neither of us but which both of us can accept as the price of agreement on other things. But it should be called a compromise, rather than insulting the world’s intelligence by calling it “non-informative”.
In any case, as a dispute about social role rather than about math itself, this can go on forever, and probably will. Of course the disputants themselves are almost all in the same social role: math professor. But they are coming up with theorems and formulas to be used by others, and the question is which sorts of others they prefer catering to. It is not, in any case, a vicious dispute; neither side is trying to exterminate the other or drive them out of academia. Thomas Dormandy has written that medical disputes are always particularly vicious when both sides are wrong. This is rather the reverse: both sides are right, at least in appropriate contexts.

Union law

Sun, 06 Apr 2025 00:00:00 +0000

I once read a description of the Code Duello, the formal set of rules that existed to govern duelling, which stated that it was a finely crafted, well-thought-out code which, in its entirety, should not exist. It seems to me the same is true today of the whole system of law that regulates trade unions and how businesses must deal with them. 
The Code Duello assumed belligerence, and tried to tamp down unfair aspects, while still leaving duellists free to kill or maim each other. It had rules for how challenges were to be given, how “seconds” were to be chosen, who had the choice of weapons, and so forth. In modern law we just assume belligerence is bad, and we severely punish aggressors, as well as letting their victims defend themselves, even if necessary with deadly force.
Union law, likewise, assumes belligerence, in the form of strikes and such, and tries to tamp down the worst excesses. It dictates that a business can’t fire someone for being a union organizer, even if he’s being a total lying asshole about it. A business can’t threaten to close a factory if the employees unionize. (They can still close the factory; they just can’t threaten to do so.)
Those, and a large number of other pro-union rules, give so much power to unions that companies hardly ever shake off their grip. As a percentage of workers in the US, unions cover something like 10% these days, down from about 30% in the 1950s; but this trend has been driven not by companies busting unions but rather by companies failing and being replaced by non-union companies. In those new companies, unionization votes often fail due to workers themselves thinking of unions as having too much power and as endangering the company’s existence. (That, of course, is if the new companies are in the US, which they often aren’t: union law has helped to drive whole industries abroad.)
But union law is not all one-sided: it imposes serious restraints on unions, too. Yale graduate students who were trying to unionize were at one point surprised to find that they couldn’t conduct a “grade strike”, withholding grades that they produced from their work as teaching assistants. The courts had to explain to them that selective strikes were not allowed, or rather not protected under union law. To refuse to do just one part of their job would be interfering with management’s prerogative to run the company. For the same reason, workers are restricted from using their power to bargain over things other than “wages, hours, vacation time, insurance, safety practices</a>” and such.
This is perhaps necessary in a system where unions are otherwise given so much power that they could just boss companies around willy-nilly if they were allowed to. But if union law weren’t excessively pro-union in the first place, for a union to strike not over wages and working hours but over corporate direction could be viewed as a selfless and noble act: rather than just trying to line their own pockets, they were risking their jobs to do what was good for the company or perhaps for society at large. Or it could be viewed as a stupid demand. But even so it would command a certain level of respect, as being a place where the strikers could actually be persuaded by a proper counterargument, rather than a place where they’re arguing their own pocketbook and no argument will suffice. (In Germany, union law takes the opposite tack, formally giving unions power in in management decisions. This may not be the best system either, but it has not destroyed German industry.)
Union law also has rules for how elections are to be conducted for union leaders. But the blowhard who is most popular with the workers isn’t necessarily the best person to run the union. Absent union law, unions could make their leadership choices in all sorts of other ways too.
Because unions still would exist even if the whole concept of a union were extirpated from the law, the way that the whole concept of a duel has been extirpated from modern criminal law. Today, being in a duel doesn’t give anyone special rights or obligations; it’s just another fight. Management-worker disputes could be the same way. Management does sometimes try to unduly exploit workers, and workers do sometimes organize to get more than they they deserve; but they don’t need to go at each other in this stylized early-1900s fashion.

An end to all this prostate trouble?

Fri, 07 Mar 2025 00:00:00 +0000

The prostate gland causes entirely too many problems. In the US, prostate cancer kills about one man of every forty. “Benign prostate hyperplasia” (BPH) is even more common, affecting most men over age 60. It pinches the urinary tract, making it hard to urinate, and is constantly in danger of transforming from “benign” to “malignant”. Even the difficulty urinating is enough of a problem that men often get surgery for it, usually TURP, a sort of roto-rooter job (except cutting through the pipe instead of just cleaning it out). 
In women, breast cancer has a similar death toll, but the breasts have an excuse: they’re much bigger; there are many more cells to go bad. They’re also much more metabolically active, capable of producing enough milk to feed a baby; the prostate’s output is tiny in comparison.
One idea I’ve seen is that the prostate serves as the body’s “gatekeeper” against sexually transmitted diseases, and in the process often gets chronically infected itself; the resulting inflammation may cause hyperplasia, first benign then malignant. Infectious causation is too often neglected these days, and sexually transmitted diseases are common, so this is not unreasonable. But it doesn’t seem like a great explanation. The prostate doesn’t filter the urinary tract; it just secretes into it; there’s no real “gate” there to be “kept”. The prostate’s position is like that of the salivary glands, which are not known for being great houses of cancer. And the epidemiology backs this up: correlations with STDs are there, but not huge. The odds ratios are between one and two</a>; in comparison, the odds ratio for human papillomavirus and oropharyngeal cancer is greater than ten. An odds ratio of ten should make the ears perk up; an odds ratio of 1.5 can be a minor effect or can be just a spurious correlation. And with the prostate, there’s not just a minor effect which needs explaining: there has to be some major-league cause.
A decade ago, Scott Alexander, in his blog “Slate Star Codex”, wrote</a>:

“About five years ago, two Israeli doctors named Gat and Goren posit the theory that benign prostatic hyperplasia, a prostate disease that affects millions of older men, is caused by incompetence of the spermatic veins. They claim they can treat it surgically, and show off rows of smiling patients with glowing testimonials. Once again, the guys are good doctors, nothing about their theory contradicts basic laws of biology, and no one else has any better ideas.
I shamefacedly admit I want this one to be true. There’s so much “well, everything is a complicated combination of genes, biomolecules, biopsychosocial stressors and immune modulators that we may never really understand” going on in medicine today that it would be super gratifying if this one mysterious disease turned out to just be plumbing going in the wrong direction. And although the prostate is about as far from my area of expertise as it is possible to be, I have to say that from a physiological standpoint their theory seems to have that rare and much-sought scientific elegance, where everything comes together in a pretty package.
As far as I can tell, the medical community has totally ignored this one. Gat and Goren have published their hypothesis and their apparent excellent results in peer-reviewed medical journals</a>. It has garnered praise</a> from prestigious figures in the field (bonus points for calling it “seminal”, especially if the pun was intentional). As far as I know, no one has attacked it or even formally expressed doubt. Yet as far as I know, it has gone nowhere.
Does everyone mutually assume that if something this revolutionary were true, someone would have noticed beyond a single article in a urology journal? Do they just decide it needs further research, and hope that this research will be conducted by someone else? Or do they think that it would end up like Zamboni’s MS cure, with hundreds of thousands of dollars wasted, dozens of unnecessary surgeries performed, and nothing to show but yet another fringe medical idea that sounded good at the time? </blockquote>
I read that some time after it was written, and found that there’s since been a small confirmatory study</a> from Germany. So it hasn’t been entirely neglected. Gat and Goren also have other papers in the area. (The latter author is sometimes listed as “Gornish”, that being the non-Hebraicized verson of his name.)
At any rate, Scott’s was more of a social view of the question than a technical one. Still, it was intriguing enough to get me to read the papers he linked to, and then to read the authors’ other papers on the subject. I found that the above language does not convey the full scope of the theory. This is not just about BPH; it’s also about prostate cancer, and also about varicocele, the top cause of male infertility. And it promises to eliminate all these problems, if caught early enough, which is not hard: screening for it is simple and cheap.
The theory here is largely mechanical; and it’s not just psychiatrists like Scott who are weak at mechanical explanations; it’s doctors in general as well as medical researchers and biologists. There is even a famous paper “Can a Biologist Fix a Radio?”, wherein the biologist author laments the unsuitability of biological reasoning, at least of the usual sort, for the task of fixing an old radio. (End result: “the radio remains broken.”) Reading medical review papers, I have often gotten quite disgusted at the way they list result after result in the fashion “X has an effect on Y”, without saying what the size of the effect is or even what dose of X is required. Especially when you want to chain two mechanisms together (X having an effect on Y and then Y having an effect on Z), you’ve got to know the numbers. The convention is that p=0.06 means “it doesn’t have an effect” and p=0.04 means “it has an effect”, but any real thought requires more than just that binary view of things.
I’m thoroughly comfortable with mechanical theory and practice, so here’s an attempt at a full rendering of the Gat/Goren theory, in language as simple as the subject permits, followed by my own comments. The basic idea is this: in healthy men, blood flows out from the testicles into the spermatic veins (of which there is one for each testicle). Each spermatic vein goes up vertically inside the body until it’s near the kidney on that side. (The kidneys are about halfway up the back.) The left spermatic vein then feeds into the left kidney vein, while the right spermatic vein feeds directly into the vena cava (the vein “as big as a cave”; it’s the largest vein in the body, leading directly back to the heart; the kidney veins also feed into it). Each spermatic vein has about seven one-way valves to prevent blood from flowing the wrong way through the vein, which it would otherwise do due to the force of gravity. The following diagram from their latest paper</a> shows the situation in healthy men. (That paper is under a Creative Commons license, which this re-use is in accordance with; I’ve slightly altered their diagram to un-abbreviate the labels. The subsequent images are also re-used under the same license.)
With age and wear and tear, the one-way valves cease to function, and blood does flow the wrong way: down the vein towards the testicles. It bathes them in poorly-oxygenated blood, which is bad for them, eventually killing the germ cells which produce sperm and causing infertility.
The medical literature often blames that infertility on the warmth which is produced by that reverse blood flow, but Gat and Goren blame the low oxygen</a>: warmth does interfere with sperm production, but it’s low oxygen that actually kills sperm-producing cells, as occurs in men with this disorder. (The warmth is still quite useful for diagnosing the disorder since thermal imaging shows it plainly.)
The blood from the testicles, still at higher than normal pressure, then continues on out the network of veins in that area, and spills into the prostate, going backwards through veins that normally drain the prostate; to use their diagram from the same paper:
Since this blood has just left the testicles, it is heavily loaded with testosterone, which promotes prostate growth. Making things even worse, that blood has a much higher than normal percentage of free testosterone. Normally the vast majority of testosterone in the blood is bound to sex hormone binding globulin (SHBG); only a tiny proportion is free, but that proportion has the actual hormone effect. In the testicular flow, there isn’t enough SHBG to bind to all the testosterone; that happens only on the way back to the heart when it is mixed with the blood returning from the rest of the body. So the total effect is to hit prostate cells with about a hundred times the normal dose of free testosterone. This makes them grow; their growth is called benign prostatic hyperplasia or prostate cancer. There’s also another effect of the high pressure: swelling the prostate (pressurizing it and expanding it). It also swells the veins in the scrotum, the “pampiniform plexus” above the testicles; when that swelling is prominent enough, it is given the name “varicocele”.
(I’d originally imagined that their reason for the high free testosterone was that the testosterone molecules took time to find the SHBG molecules and bind to them, but that process seems to be too fast: this study</a> found a half-life of 12 seconds for dissociation, and at equilibrium the rate of association must equal the rate of dissociation. So overwhelming the local supply of SHBG seems like a better explanation, though being out of equilibrium might add to it a bit.)
This theory can explain why giving men testosterone doesn’t seem to increase the risk of prostate cancer, despite testosterone promoting prostate cancer. Giving men testosterone shuts down their own production of testosterone, making backflow from the testicles to the prostate harmless. Thus even with more testosterone in their blood, they can have less in the prostate gland.
It can also explain why low testosterone is correlated with prostate cancer: the backflow damages the testicles via hypoxia, lowering testosterone, while simultaneously funneling the testosterone that still is released directly to the prostate.
The way Gat and Goren fix this is simply by destroying the spermatic veins. Under fluoroscopy (that is, viewing an X-ray image continuously during the procedure), they snake a catheter in through the veins until it gets to the top of the spermatic vein. They then inject a pulse of X-ray contrast agent to make the flow in that vein visible under the fluoroscope, proving that the flow really is going the wrong way. Having done so, they then have the patient close off the bottom of the vein with finger pressure while they inject a sclerosing agent into the vein; they start by moving the catheter to the bottom and sclerosing that portion, then repeatedly withdraw the catheter a bit (moving up) and inject more. After a few minutes the vein is thoroughly and permanently clogged. This is their diagram of the results:
To say a bit more about the background of this theory, venous circulation is not a one-path-only thing: the veins are not like a tree, where there’s only one path from the trunk to each leaf. Indeed, this multiplicity of paths is why the problem arises in the first place: the testicles being drained both by the spermatic veins and by the local circulation. It’s also why it can be solved in this manner: the remaining veins enlarge a bit to accomodate the flow, as well as new veins growing. But it’s also why they need to be thorough when performing the procedure. Besides destroying both spermatic veins (a weakness, Gat and Goren say, of other doctors’ work, where it’s often only done on the left vein where the swollen veins tend to be more prominent), they also follow up by injecting more pulses of contrast agent to make sure that there’s no collateral circulation bypassing the blocked vein, and that the veins they did treat are really blocked. (The finger blocking the vein at the bottom keeps the sclerosing agent where it belongs rather than letting it spread further into the body. Of course a little of it still spreads, but the sort of agent they use is harmless at low concencentrations.)
It’s odd for there to be such an easily-removable design flaw in the human body; evolution tends to remove them. Since it strikes at advanced ages, BPH doesn’t make a big impact on a man’s ability to pass on his genes. But being the leading cause of male infertility sure does. Their explanation is that evolution hasn’t had much time to work on the problem; in animals the spermatic vein is horizontal, and doesn’t have or need one-way valves. It’s our standing upright that yields the problem; in evolutionary terms that’s a recent development.
So that’s their theory. There is an obvious question, though, which mechanically-inclined reviewers might have raised, and which deserves an answer. It has to do with pressure and height. As Gat and Goren say, it is a law of physics that pressure within a fluid increases as one descends deeper into the fluid, in direct proportion to the depth. (They invoke Bernoulli’s equation, which also involves the velocity of the fluid, but they then assume the velocity is negligible, so are using only the part of the equation that relates pressure to height.)
But they apply this law only to the blood in the spermatic vein. It is also applicable to the blood in the arteries and to the blood in other veins. Doing so, though, would ruin their argument: it would mean that although pressure increases as one descends down the spermatic vein, it also increases in the other veins. In their account, blood goes backward through the spermatic vein, emerging at a pressure of 40 cm H2O or so, then pours into other veins which, drained by the common iliac vein which in turn is drained by the inferior vena cava, are at less than 10 cm H2O. (They use mm Hg as their units, but I’ll be converting all pressures to cm of water, since it’s a more directly relevant unit: blood’s density is close to that of water.)
But gravity acts also on the iliac vein and the vena cava! In general, if you have a loop of tubing in a gravitational field, and the fluid inside is not in motion, the pressure of the fluid is everywhere proportional to altitude, no matter how the tubing is shaped. No spontaneous circulation arises. Here we have a loop with the spermatic vein being on one side and the vena cava and iliac vein on the other, connected on top and bottom, and the claim is that this is enough to give rise to spontaneous circulation. It’s not.
Yet there is reverse flow in the spermatic vein when the valves fail; that’s standard medical knowledge. And, as mentioned, Gat and Goren confirm the reverse flow by injecting tracer dye into each and every vein they treat, before sclerosing that vein. The reverse flow exists; it just has to have a more complicated explanation than the one they provide.
Another part of their argument is that venous blood infiltrates the prostate due to the venous pressure exceeding the local pressure in the arteries that feed the prostate. But gravity also acts on the blood in arteries. If the local arterial pressure were equally boosted by height, it could not be exceeded.
Fortunately for their argument, that’s not the way the human body works. It sort of theoretically could work that way, if all the blood vessels were rigid: pressure everywhere would increase with decreasing altitude in the same way that pressure increases with depth inside the ocean. Relative pressures would be higher in arteries, lower in veins, and intermediate in capillaries: the pressure difference between artery and vein would be the same throughout the body, but the pressures of both would be highest in the feet and lowest in the head. (For fish, or babies in the womb, this works even without blood vessels being rigid, due to the outside pressure varying with altitude in the same way as the pressure in their blood vessels does; all the fluids involved have densities close enough to the density of water that the difference doesn’t matter.)
Of course in fact blood vessels are not at all rigid. Arteries are the closest to being rigid; they’re thick-walled and can take a lot of pressure, though they still have an important amount of elasticity. Veins, though, are thin-walled: they swell up with higher pressure and collapse flat at negative pressure. Capillaries don’t swell, but at higher pressures they leak fluid into the surrounding tissues, producing swelling there. So the body maintains capillary pressures in the neighborhood of 20 cm H2O (somewhat higher at the start of the capillary and somewhat lower at its exit).
The usual figures for blood pressure (the ones for which “normal” is about “120 over 80”) are arterial pressures in units of mm Hg, measured with a blood cuff around the arm (at about the same altitude as the heart). In that example, the 120 is the maximum pressure just after the heart beats, and the 80 is the minimum pressure it reaches between beats. Those numbers translate into about 160 cm H2O and 110 cm H2O.
In a standing person, the arterial blood pressure is considerably higher if measured at the ankles, as it would be in the rigid-vessel model. But then as large arteries divide down into smaller and smaller arteries, most of that pressure is spent before it gets to the capillaries. This is not just a matter of passive friction but also of active control: arteries have muscular walls and contract to restrict blood flow to the part of the body that they feed. When more blood flow is needed somewhere, the local arteries relax to let it through.
Veins operate at low pressure, usually less than 15 cm H2O; the hard part, and the part that is most relevant here, is explaining how they get blood upwards against gravity, since their pressure is nowhere near enough to do so. The one-way valves in the spermatic vein have already been mentioned, and are present in many other veins too, but valves themselves don’t do any actual pumping work (in the physics sense of the word “work”). Yet work needs to be done to pump blood uphill. In most cases that work is supplied by muscle movements in nearby muscles, which compress the veins. In engineering, pumps are often built with two one-way valves, the volume between those valves being alternately compressed and expanded; the principle is the same here. But since it relies on muscle movements which are made for other purposes, this process is slow and uncertain. So to get the same throughput the veins are much wider than the arteries.
When the veins’ one-way valves fail near the surfaces of the legs, it results in high pressures and swollen, unsightly “varicose veins”. Even with functioning valves, standing motionless results in high venous pressures in the feet; thus the advice, on long trips sitting down, to take breaks and to move the feet and legs around, so as to avoid blood stagnating and clotting (a “deep vein thrombosis”). For blood circulation fidgeting is good, even if it’s not “good manners”.
But the vena cava is different: it doesn’t have any valves; neither does the iliac vein (at least the portion of it under discussion here). Yet somehow blood which feeds into it at low pressure manages to climb to the heart, overcoming that 40 mm H2O pressure difference and more.
I looked in the medical literature for how this actually happens, and was disappointed not to find a clear, definitive answer. It’s common to see it stated that breathing has an effect. The heart is located between the lungs, so shares a common pressure compartment. The pressure outside the heart is about the same as the pressure outside the lungs. That pressure decreases to inhale and increases to exhale. But those pressure excursions are minor: about 1 cm H2O in either direction. (This rises with heavy breathing, as in exercise; but the blood circulation has to work even when the body is calm.)
There is also a disturbing level of confusion in the medical literature as regards what “zero pressure” is. Good engineering practice is to say what reference pressure one is measuring against. Commonly that is the ambient pressure (the pressure of the air surrounding the body, which is about 1000 cm H2O at sea level and lower at higher altitudes). To inhale, one obviously has to drop the pressure in the lungs below ambient pressure. A measurement relative to ambient pressure is called “gauge pressure” by engineers, because it’s what the usual sorts of pressure gauges tell you. The old-fashioned mercury manometers for measuring blood pressure measured gauge pressure, since they were open to the air on their other end. Modern mercury-free manometers are no doubt built to give the same numbers (so as to be compatible, and also because it’s easy to build a gauge that way.) So the usual arterial blood pressure numbers are all referenced against ambient pressure.
That is so even if medical literature tries to confuse the matter, as happens. Take, for instance, this passage in Guyton and Hall’s Textbook of Medical Physiology, 14th edition (a standard text in medical schools):

Pressure Reference Level for Measuring Venous and Other Circulatory Pressures Although we have spoken of right atrial pressure as being 0 mm Hg and arterial pressure as being 100 mm Hg, we have not stated the gravitational level in the circulatory system to which this pressure is referred. There is one point in the circulatory system at which gravitational pressure factors caused by changes in body position of a healthy person usually do not affect the pressure measurement by more than 1 to 2 mm Hg. This is at or near the level of the tricuspid valve, as shown by the crossed axes in Figure 15-12. Therefore, all circulatory pressure measurements discussed in this text are referred to this level, which is called the reference level for pressure measurement…
When a person is lying on his or her back, the tricuspid valve is located at almost exactly 60% of the chest thickness in front of the back. This is the zero pressure reference level for a person lying down. </blockquote>
Now, what is meant by that? It purports to define a pressure reference level, and one that is not just equal to ambient pressure. But does it mean that every single blood pressure measurement in the entire book was done by snaking a catheter into the heart until it reaches the tricuspid valve, and using that as reference? Of course not! That would be ridiculously dangerous and expensive. Also, that valve is in the middle of the heart: it’s the valve between the right atrium and the right ventricle. The pressure there varies strongly as the heart beats, making it unsuitable as an experimental reference pressure.
So the passage cannot really mean that it is defining a reference pressure. Instead it is defining a pressure reference altitude. I take it to be just an overly fancy way of standardizing the practice of putting the blood pressure cuff around the top of the arm with the arm draping normally down the side of the body, which puts the cuff at about heart height. (Even on someone lying down, that position of the cuff is still at about heart height.)
Unfortunately it also betrays a lack of real precision, which is probably unimportant as regards the usual use (arterial pressures), since those numbers are large, but is important when dealing with the much lower pressures in veins and the issue of how venous blood is pushed back from the lower body to the heart.
This lack of a definition also extends to the Gat and Goren papers: they give pressure numbers, but never say relative to what. Ambient pressure generally seems like the best reference pressure to use; I will be hoping that papers I cite for pressures used it, but also, since none of them say they do, not placing much trust in any absolute numbers.
Enough complaining; time to return to the question of how blood makes it back to the heart. As mentioned, inhaling a breath reduces the pressure in the heart and lungs; but that reduced pressure ends at the diaphragm, the sheet of muscle and tendon that separates the thorax (heart and lungs) from the abdomen (guts). The diaphragm is arched upwards; to inhale we contract it, flattening it and lowering the pressure in the heart and lungs while increasing the pressure in the abdomen. Those two pressure changes add together to force blood in the vena cava up past the diaphragm toward the heart. (This is for normal breathing; in general the pressure in the abdomen can vary quite considerably depending on what the abdominal muscles are doing, as measured for instance in this</a> paper. Above the diaphragm, it also depends on what the rib muscles are doing, they being another force that powers breathing. But in normal breathing the diaphragm is the main actor.)
So breathing helps as regards the general problem of returning blood to the heart. Still, the heart does need blood feeding it when breathing out as well as when breathing in. We don’t synchronize our breaths with our heartbeats, nor could we, since breaths are less frequent. (The frequency of the heartbeat does change a bit in breathing out compared to breathing in, though.)
It helps that the lungs are elastic, and would collapse if there weren’t more pressure inside them than outside. This means the normal pressure in the thorax is negative, helping to pull blood in. This “transpulmonary pressure” (the difference between the air pressure in the lung and the fluid pressure outside it) is about 6 cm H2O.
The heart, as it relaxes from its contraction, presumably also can exert a bit of suction on the incoming flow; it’s obviously not rigid enough to exert much suction, but every little bit helps.
That the system operates right on the edge, getting blood back to the heart with almost no pressure to spare, can be seen by the reader if he examines the veins on the back of his hand: if held at lap level those veins stand out, full of blood, while if raised to eye level they flatten. By moving the hand up or down slowly, giving the veins time to empty or fill, the level where the transition from full to flat occurs can be determined; it’s about at the same level as the heart. (This demonstration probably won’t work for everyone; blood vessels are less prominent in the young, and also can be hidden by fat.)
Now to return to the main question here, which is backflow in the spermatic veins. Those veins are entirely below the diaphragm. They are in the same pressure compartment as the vena cava is at that altitude (the retroperitoneal space), so the two competing flow paths are under about the same pressures.
I’m afraid that the answer is that Gat and Goren are simply wrong about the pressure at the bottom of the vena cava when standing: it must be at considerably higher pressure than the sub-10-cm H2O that they say it is – or at least it is when the patient is standing up (or sitting normally). It’s an understandable error, since the medical literature usually refers to the lower numbers: in circumstances where it matters, patients are commonly horizontal.
I found a paper from 1966</a> which measured intravenous pressures both with the patient horizontal and with the patient standing: the horizontal numbers are in accordance with Gat and Goren’s, but the standing numbers are considerably higher. Using their numbers for the common iliac vein (since they don’t say where in the vena cava they measured pressures, and since the common iliac vein is where the veins from the prostate enter, so is the most relevant number anyway), the average of their three measurements has the pressure with patient horizontal as 8 cm H2O (plus an additional 9 cm H2O when breathing in), and the pressure with patient erect as 30 cm H2O (plus an additional 15 cm H2O when breathing in). If we accept these numbers, the pressure in the common iliac vein is close to the pressure in the bottom of the spermatic veins.
Again, though, it’s generally agreed that the backflow exists, at least through the spermatic vein. (Gat and Goren’s contribution is extending it to the prostate.) The large pressure difference they say is driving the backflow is calculated based on improper assumptions, but there must be some pressure difference driving the backflow.
Now, one difference between the two veins competing for flow here is that the vena cava is much larger and thus has much less frictional resistance to flow. In general, blood flow in the body is dominated by frictional effects (fluid friction, that is, aka viscosity): Gat and Goren cite Poiseuille’s equation for fluid flow, which states that flow is proportional to the fourth power of the diameter of the pipe. In comparison, if friction were zero, the flow would only be proportional to the square of the diameter of the pipe (that is, proportional to its cross-sectional area). But Poiseuille’s equation, which has it as the fourth power, is the generally-accepted formula for blood flows. It’s just an approximation, not a fundamental law, but the fact that it’s the accepted approximation shows how dominant friction is in blood flow.
The ratio in diameters between the spermatic vein and the vena cava is about a factor of 7. So if equal lengths of the two veins are subjected to the same pressures, Poiseuille’s equation says that the vena cava has 2401 times the flow and 49 times the average flow velocity. But that just predicts much less forward flow in the spermatic vein; it does not predict that it backflows.
Yet it does. The hydrostatics don’t predict it, though; we need to look into its hydrodynamics. The two main sources of dynamic behavior (pulsations or fluctuations rather than constant flows and pressures) have already been mentioned: one is the right atrium, alternately accepting inflow then contracting in the heartbeat, its input valve slamming shut to prevent backflow. The other source is the breathing.
Since intestines have basically nothing holding them in place (as can be attested to by any hunter who has gutted an animal, slitting open its belly and seeing the guts spill out), it’s not just the blood in the veins but the whole abdominal compartment that is reasonably well modeled as having a pressure gradient determined by altitude: about one cm H2O pressure for every cm of altitude (since most things in the body have a density close to that of water). Of course pressures inside a pressurized cavity (such as the bladder, a vein, or an intestine) can be higher, but even within those cavities there is the same pressure gradient with altitude (unless filled with air or some other gas, as intestines sometimes are).
Though I earlier alluded to the retroperitoneal space as a pressure compartment, the peritoneum is a thin membrane which, in surgery, tears easily</a>; there’s not much chance of it holding significant pressure for long. What mainly holds in abdominal pressure are layers of muscle and connective tissue. In the bottom of the abdomen this is the pelvic floor muscles. In men, those are just below the prostate, so the prostate too is in the same pressure compartment.
What isn’t in that pressure compartment are the testicles; they are hanging out in the breeze, surrounded by ambient pressure.
The level of influence that the heart’s beating has on the pressure in the vena cava can be judged by the velocity of flow in it. According to measurements done in this paper</a>, the velocity in the inferior vena cava peaks at 45 cm/second (in people breathing normally while lying down, and measured at the level of the T11-T12 intervertebral disc). Per Bernoulli’s equation, that energy density is the same as is produced by a pressure rise of 2 cm against Earth’s gravity. (This is also the height an object rises to if it is thrown upwards with that speed.)
2 cm H2O is not much pressure rise, particularly compared to the above-cited 15 cm H2O pressure increase in the abdomen for breathing in. Also, the pressure pulse from the heartbeat is entirely inside the veins (and diminishes with distance from the heart), whereas the 15 cm H2O pressure increase is applied to the whole compartment uniformly. (But neither of those numbers is particularly certain, and both must vary considerably depending on circumstances.)
As mentioned above, capillaries and veins can only hold limited blood pressure. But what matters is the difference between external and internal pressure. If capillaries exceed maybe 40 cm H2O, they leak fluid, producing edema; but inside a pressure compartment at 30 cm H2O that internal pressure limit increases to 70 cm H2O. Numbers like those occur near the bottom of the abdominal cavity when standing: pressures higher all-around without producing distress. But if the testicles, which are outside that pressure compartment, experience the same internal pressure, that will produce distress in capillaries and veins.
While intact, the spermatic vein is protective: its one-way valves mean that blood is drawn upwards when the abdominal pressure is at its minimum and can’t return when it’s at its maximum. Still, to get into the abdomen at all, the blood in the spermatic vein has to be at least at the pressure of the abdomen (at the point of entry, which here is near the bottom of the abdomen). The abdominal pressure varies; if the pressure in the spermatic vein is between the minimum and the maximum abdominal pressures, then the part of the vein which is in the abdomen will be collapsed part of the time, which with intact valves makes for effective pumping. If the pressure in the spermatic vein is above the maximum abdominal pressure, it doesn’t need pumping to flow, but still functions as a pump, via the vein expanding and contracting.
So when the one-way valves fail, the pressure at the bottom of the spermatic vein increases. Still, the difference in pressure between the failed and working states is not the full 40 cm H2O that Gat and Goren deduce from the height of the spermatic vein, but rather cannot be more than the 15 cm H2O pressure difference between breathing in and breathing out.
Now, that last number is just a measurement from one paper, likely from one single Irishman; and if he had been breathing harder the number would have been higher. So it should not be taken as gospel; but it wasn’t easy even finding that number. Not that I pretend to thoroughness; I could well have missed a better source, and readers who know of one are invited to provide it. But I’ve spent enough time looking that at this point it seems like it might be easier to measure intra-abdominal pressure changes myself, sticking a pressure probe up my butt – that’s literally how this paper</a> did it (but on horizontal patients rather than vertical ones, so their numbers are not quite on point here.) That’s not quite the same thing that the Irish paper measured, which was intravenous pressure; sticking a catheter into my veins is way beyond my skill set. But for measuring pressure increase it’s close enough, since the pressure increase from breathing is applied to the whole compartment.
In any case, that’s why there’s higher pressure at the bottom of the spermatic vein when the valves fail, but it doesn’t answer the question of why there is backflow. The situation, again, is of two vertical veins, connected at top and bottom, in the same pressure compartment, with one vein being much larger than the other. Static, constant flow would just be divided between the two; both would flow upwards. But consider rhythmically varying flow, such as is produced by the heartbeat or by breathing. First there is an acceleration phase, which acts on both veins. In the vena cava, with much less friction, the blood accelerates upwards, whereas in the spermatic vein, with more friction, there is much less acceleration; the extra upwards force is mostly spent against friction. Then there’s a deceleration phase. The blood in the vena cava has inertia; it’s trying to go upwards but has to stop, so the pressure at its top increases. Some of that higher-pressure blood then is pushed down the spermatic vein. And the flow up the vena cava has such a dominant share of the overall flow that even a slight influence like this can yield net backflow in the spermatic vein.
Now, the mechanism I just described does not seem like a particularly strong mechanism; but as mentioned, in practice it’s well established that there is backflow; this is just one way that theory can agree with that observation. There may be other mechanisms that do better, perhaps involving the portion of the flow which is outside the abdominal pressure compartment, and/or the relatively narrow veins communicating between the two paths at the their bottom – though that gets complicated enough that it’s not easy to think through; a computer model might be in order.
More questionable, although still quite plausible, is the idea that the backflow reaches the prostate. The Gat and Goren theory is that it is pushed by about 40 cm H2O pressure, overwhelming even the prostate’s arterial supply. But that’s pure theory; they don’t report actually measuring any pressures. (If they were researchers I would ding them harshly for not doing the measurements, but they’re clinicians; they’re paid to cure patients, not to produce knowledge.)
Gat and Goren didn’t do a great job of proving experimentally that backflow reaches the prostate, either. They measured the high free testosterone levels described above, but that measurement was at the bottom of the spermatic vein; they didn’t have their catheter make another turn and go into the deferential vein and then past its junction with the vesicular vein and into the vein coming from the prostate, so as to measure high free testosterone levels in a place where they definitely don’t belong. (This is readily excusable: those other veins are much smaller, so the catheter might not fit or might not be able to make the sharp turn into the smaller vein. Again, researchers should be dinged for omitting such tests – there’s got to be some way to measure pressures there – but clinicians get a pass.)
Perhaps the strongest piece of evidence they offer for backflow to the prostate is an X-ray showing contrast agent which was injected into the spermatic vein. They state that:

On retrograde venography of the PP, after a delay of about 10 s, a contrast material ‘blush’, of the prostate gland capsular region, was observed (Fig. 3). Both are clearly seen in the image. </blockquote>
Here’s their Figure 3 (again, reproduced from their most recent paper under the terms of a Creative Commons license):
Looking at it, I can see the oval they label “prostate”, but it’s pretty faint. There are various other structures (such as the bladder) which are also visible with similar intensity. Backflow down the spermatic vein is clearly visible (all that darkness going down to the region marked PP, for “pampiniform plexus”, the network of blood vessels above the testicle), but backflow down the spermatic vein is not the new part of their theory. There is some darkness under the the prostate, which looks like backflowing dye, but is not clearly connected to the dye in the PP region. Even if it is dye, it’s not evident whether it gets into the prostate itself. Perhaps specialists who are used to viewing fluoroscopy images will pronounce this one definitive, but for me it’d take a movie showing contrast emerging from the tip of the catheter and then making its way to the prostate; in this day and age, movies should be easy to provide. Of course in their clinical practice they are seeing such a movie with their own eyes for every patient, so it requires a modest level of distrust to doubt them on this. But it’s not entirely unknown for doctors to succumb to wishful thinking, and “blush” is not a strong word. I’m not inclined to distrust them on this, but the image they show is less than a complete proof.
I’d also be curious, with such a movie, to look for variations in the flow. There would no doubt be motion visible from breathing, and the contrast agent would show how that correlated with blood flow in the veins.
Breathing would help spread the testosterone-loaded outflow, since the prostate is inside the abdominal pressure compartment while the testicles are outside. Breathing would thus push the flow alternately in one direction then the other, increasing the chance of subjecting the prostate to outflow from the testicles. The movie would show this. Even if it resulted in only a weak and diluted flow which didn’t deliver the theorized hundred-fold increase in free testosterone, a ten-fold increase is still quite a lot and even a doubling would be important.
A really thorough experimental workup would measure all the relevant flows and pressures, perhaps by using an intravenous catheter loaded with multiple pressure and flow sensors each delivering many measurements per second. (Whether such a catheter can be bought today and whether it can be readily inserted into the veins in question are not matters I have looked into, but with modern technology there’s got to be a way of measuring such pressures.) Testosterone levels would also be measured at various points.
Recently a paper by Alyamani et al</a> came out which did do such measurements of testosterone. The patients were 266 men undergoing prostatectomies for prostate cancer, none of whom had any prior hormonal therapy. During the course of those operations the researchers extracted blood from the prostatic dorsal vein and tested it for testosterone. They also tested the removed prostates for testosterone. They indeed found levels of testosterone in the dorsal vein and in the prostate which were many times higher than the levels measured in normal peripheral blood from the same patient. Related substances in the samples were consistent with the blood having come straight from the testicles. They dubbed this phenomenon “sneaky T”, and describe their results as supportive of the Gat/Goren theory.
If I were just being booster-ish, I could leave it at that (“yay, support”); but I’m taking a hard numerical look. And from a numerical point of view, at first glance that support seems weak. Only 20% of the men had dorsal vein / peripheral vein ratios of testosterone which were two or greater. About half of them had ratios below one. This is not what you’d expect if this were the main driving force of prostate cancer: you’d expect seriously elevated ratios in pretty much all the men, not just in 20% or 50% of them.
The association they found between dorsal vein testosterone and prostate tissue testosterone is also weak. By biology standards it’s totally there (P = 0.004), but there were still a lot of men with high levels in one of those places and low levels in the other. This is not what you want for the definitive mechanical cause of something.
But the title of a paper</a> they cite caught my eye: “Testosterone and estradiol are co-secreted episodically by the human testis.” Looking at it, I find it could explain the above weaknesses: the testicles don’t put out testosterone all the time, but rather do so only in pulses, at the rate of about one pulse per hour. Testosterone stays in the bloodstream long enough that testosterone measurements in circulating blood don’t vary much, but the output of the testicles varies quite a lot. The paper’s measurements (once per fifteen minutes) weren’t frequent enough to really pin down the length of each pulse, and accordingly they give no figures for that length; but eyeballing their graphs, maybe elevated testosterone is present about a quarter of the time, though it’s hard to tell.
So when the Alyamani paper found doubled levels in 20% of men, that could mean that it was actually happening in all of them, just only about 20% of the time in each. Or it might be happening in half of them, about 40% of the time in each. Also, the stress of surgery might have affected the timing of the testosterone pulses.
The pulsing also could explain prostate tissue concentrations often being different from dorsal vein concentrations: testosterone takes time to sink in, so if they measured on the leading edge of a pulse the tissue concentrations would be lower than the vein concentrations, whereas on the trailing edge they’d be higher. Or the measurements might have been done at different times during the course of the surgery; the dorsal vein blood might have been extracted, say, ten minutes prior to the prostate’s extraction.
So the Alyamani paper’s measurements are consistent with backflow of high-testosterone blood being a factor not just in 20% or 50% of cases but in all of them. Still, the backflow, when present, is doubtless stronger in some men and weaker in others. And indeed being in that top 20% did have some predictive value: prostate cancer recurrences were more common in the 20% group. (Since the prostates themselves were all removed, this presumably represents metastases. It would be interesting to know whether the increased metastases were in the same local area bathed by high testosterone or whether they were distant; but the paper does not report this.) They also did not measure free testosterone, just total testosterone, so that’s another unknown.
Besides the BPH paper, Gat and Goren even tried their method on outright prostate cancer</a> (early stage, Gleason 3+3) and report disappearance of cancer (as measured via biopsy) in 5 out of 6 patients, along with lowering of PSA levels.
One thing to note, though, is that their fix of destroying the spermatic vein and any collaterals can’t make the system as good as it was originally. Destroying a backflowing vein always provides improvement in pressures (since otherwise the vein wouldn’t be backflowing), but it can’t improve pressures as much as forward flow would. The ideal would be valve repair or replacement; that is done routinely on heart valves, but not on vein valves. (This also explains why evolution hasn’t simply removed the spermatic vein in us: when it’s there and working, it helps.)
Operations for varicocele are done widely, to try to fix male fertility, but aren’t always of this sort: it’s common to see the swollen veins in the scrotum as the problem and simply remove many of them surgically. This does not squarely address the problem of backflow, and indeed there’s a bit of debate in the medical literature over whether it produces statistically significant fertility improvements. (On the whole it seems to, but there are enough failures to produce debate.)
Even among doctors using the embolization approach (that being the term for inserting a catheter and using it to apply a sclerosing agent), many do it only on the left side, that usually being the one with the visible varicocele. (There is an asymmetry among the spermatic veins, with the left vein being a bit longer and thus more problematic.) Besides doing both sides (or at least checking both sides for backflow with contrast agent), sclerosing each from top to bottom, the full Gat/Goren procedure is also to check for collateral veins and sclerose those too, and then follow up with more contrast agent to ensure that it was done properly.
So there’s a chance that a varicocele surgeon will do this properly, but one can’t just go to a random varicocele surgeon and assume he’ll do it properly. Even someone who does believe the Gat/Goren theory can screw it up in practice, especially if you’ve just got done persuading him and this is his first try.
Screening for this disorder is simple: use a thermal camera and compare testicular temperature sitting up (or standing) versus lying down, in each case waiting five minutes or so for temperatures to equilibrate, and taping the penis up so that it does not affect the measurement. According to Gat and Goren</a>, this is almost as good a test as measuring backflow on fluoroscopy, and a far more sensitive test than trying to look for the enlarged veins of varicocele. (Rather than a thermal camera, they use a liquid crystal strip which is designed and branded for this exact job; but given how expensive medical items are, these days thermal cameras are likely cheaper and are certainly more widely available. Their calibration is probably not as good, though.)
In the above-referenced blog post, Scott Alexander goes on to write that neglect of this sort of new idea isn’t some sort of conspiracy; it’s just the default. For starters, most new ideas everywhere are wrong and deserve to be neglected. And here it’s not like drugs where the patent system incentivizes people to develop new ones and gets them enough money to test them thoroughly and then pay for billions of dollars in advertising. It also is not one of the “hot” fields in science where scientists are racing to study the possibilities. It’s not genomics or the microbiome or any other “-ome”; it’s just plumbing.
He might have added that most of society’s attention on medicine is focused on squabbling about who pays for it, with precious little attention given to possibilities of finding new ways to actually cure disease. Yet those can represent the most dramatic cost savings; here, a thousand-dollar procedure might mean not spending hundreds of thousands of dollars on cancer. The thinking by the powers-that-be should not always be “oh, no, not another thing that we’ll have to spend even more money on”, even when that is true in the short run.
Malpractice law also stacks the deck against innovation: it doesn’t blame doctors for doing what doctors generally believe in doing (the “standard of care”), even if the outcome is bad, but it does blame them in the event of a bad outcome from something new and innovative. (Which in a way it should, since most new ideas are wrong; but it’s never really fair to put only the failures in front of a scientifically ignorant jury and invite them to shell out millions in damages.)
Insurance doesn’t pay for “experimental” treatments, either.
Advertising by doctors is also strongly frowned-on, particularly advertising that goes against conventional wisdom; a doctor who blanketed the local airwaves with claims of a prostate cancer cure (or even a preventative) would very likely get his license yanked by the state medical board. (Not that broadcast advertising really would be the way here: too costly for an individual doctor. The modern way would be targeted Facebook ads, relying on Facebook’s intimate knowledge of personal medical problems. The medical board might not even find out about it, though if they did they likely would be extra wrathful.)
The back-to-nature crowd is also not interested, at least not in the financial sense of “interested”, because there’s nothing here they can sell as a “dietary supplement” and they aren’t permitted to perform medical procedures. Perhaps they could advocate going back to walking on all fours? Nah, too impractical. Going back to the oceans and living like fish, with a supportive pressure gradient all around us? Nah, even more impractical (though there already is a sect of “aquatic ape” believers out there).
None of this amounts to conspiracy – it’s all in the open – but it adds up to quite a set of obstacles. (It’s not like anyone’s paying me to write this, or like I have much prospect of making money from it; it’s just the strong curiosity of someone whose grandfathers both died from prostate cancer.)
Being surgery, this procedure doesn’t need to go through FDA approval, so there’s no decade-long process to form an obstacle to people getting it. Finding a doctor who will do a good job of it, though, is not simple; a short glance at patient forums reveals considerable frustration. I can’t tell if the Gat/Goren clinic in Israel is still open for business: their website www.pirion.co.il isn’t responding, and given those doctors’ ages it wouldn’t be surprising if it has closed. The German clinic which did the 2014 reproducing study</a> might still be doing the procedure. At any rate, though, a single clinic can’t handle worldwide demand for such a common disorder – not even just elite demand.
The procedure also doesn’t last forever, according to Gat and Goren’s latest paper</a>. They write that new venous bypasses grow to replace the destroyed spermatic veins: at first these are tiny, but then grow to where the problem recurs. Part of their reasoning, though, is faulty: they write that at first the size of each growing vein is small enough that capillary action can draw fluid up. Now, it’s quite true that capillary action (the physical/chemical attraction between the fluid and the walls of the capillary tube) can draw fluid up through great heights; trees, for instance, rely on it to draw water into the treetops. But it can only draw fluid up into empty space; it has no lifting effect when the tube starts out completely filled, as is the case here. (In trees it’s the evaporation at the top that powers it.) A simpler explanation is just that those tiny veins are too small to cause the problem to recur. In any case, the paper makes no comment as to whether the problem can be solved the same way a second time; obviously in principle it can, but finding all the new bypasses and sclerosing them might be difficult in practice.
As for mass adoption, I’m well aware that this essay is not the stuff of a mass patient movement. I’ve tried to make it accessible, but the needs of actually being accurate and making a thorough technical argument mean that not more than one in a thousand patients would read to the end of it. (If you think otherwise, please consider that your social circles are not typical.) A video with animations would improve that proportion considerably, maybe to one in a hundred, and I’ve considered it, though I’m not handy with animation tools. But it’ll take serious interest from scientists and doctors to really put an end to all this prostate trouble.

Academic freedom

Sat, 16 Nov 2024 00:00:00 +0000

Academic freedom has always seemed a weird concept to me. I mean, universities generally promise it to their professors; I’ve seen many people argue for it; I’ve hardly seen anyone argue against it (at least not in principle – it’s always “this is an exception” rather than “the rule is wrong”); but still, the whole idea seems so naive. When someone has been hired as a world-class expert, has been entrusted to teach others, and then reveals himself to be a prize ass, how can anyone look at him the same again? 
It goes the other way, too. Lots of professors these days have been hired to be prize asses. If they suddenly start making sense, they can’t really expect any mercy from their employers, no matter how much lip service is paid to academic freedom.
What does make sense is to give professors the benefit of the doubt. If there’s doubt about what they meant, or whether they were just joking, or whether what they meant is actually wrong, they should get the benefit of that doubt. Giving people the benefit of the doubt is generally a good rule, except in adversarial situations such as when someone is threatening you or trying to sell you something, and it especially makes sense when judging experts whose job is often to say something that’s beyond the average person’s comprehension. It is good tactics even within academic subjects that are bullshit: how do you expect to do a good job propagating your bullshit if you keep crucifying your top bullshitters for occasional mistakes?
Of course the benefit of the doubt only goes so far; after a few iterations of it, there isn’t really much doubt left. Two coin flips in a row ending up tails is normal; ten in a row isn’t.

The Wire

Thu, 07 Nov 2024 00:00:00 +0000

The Wire is one of the great shows of our time: exciting yet sticking close to reality. I generally disdain TV shows and movies, and have hardly even heard of most of them. But The Wire has the reputation of being highly realistic, which is what got me to watch it in the first place, and it didn’t disappoint. It doesn’t get absolutely everything right, though. 
One of the big things it gets wrong is the conduct of top-level gunfighters, of which it has two: Omar Little and Brother Mouzone. There’s no mystery as to why: as they say explicitly on the DVD commentaries, David Simon and company were imitating Westerns; and Westerns are profoundly unrealistic.
The first scene at which I really winced was when Omar stands with his back to a door behind which there are drug dealers and demands that they surrender their stash – which they then do, dropping it from a window above. I don’t care how good a gunfighter you are; that’s just not a viable tactic. Yes, good gunfighters do often get their way without having to shoot, but that’s usually because they put themselves in a position of total advantage where it’s obvious to their opponents that they have been outmaneuvered and that resistance is useless. Indeed, that happens in other scenes: somebody suddenly finds himself looking down the twin barrels of Omar’s shotgun. But here Omar is at every disadvantage. Entering a house and clearing it is an almost impossible task: there are too many places someone can wait in ambush. There are tactics for house-clearing, and militaries do practice and use them, but they do so for occasions where there’s only something like a one-in-a-thousand chance that an enemy is waiting in ambush. When they expect resistance, their tactics are very different and involve a lot more use of high explosives. “Doors are walls and walls are doors” is a saying from that setting, meaning never to go through a door because someone will be waiting to shoot you: instead set an explosive charge and make yourself a new door in a wall. Or you might call for air support and level the whole building with a bomb, or at the very least throw a lot of hand grenades around corners.
But Omar doesn’t have an air force or an army or anyone at all backing him; he’s alone against a whole gang. And then on top of that he turns his back. Well, given the risk of being attacked from behind, maybe that actually would be the best use for his eyes, while his ears attended to the house. But he isn’t even using his eyes to scan for threats: they’re downcast most of the time. He’s not even doing the relaxed awareness thing (where you let your eyes relax and rely on peripheral vision to pick up any movement).
Even if he were using his eyes to good effect, it would just show how bad his overall situation was, with threats that might come at him from anywhere in 360 degrees – plus above, where they could have sent down a bullet rather than the bag of drugs. They threaten him by saying they have a MAC-10, which he disbelieves – reasonably so, since a MAC-10 is a big chunky machine pistol that is hard to conceal. Still, to believe that they’re certain to be completely unarmed is unwarranted: the show often has the gang members flashing pistols in that vicinity.
Now, an author does have the privilege of making his characters do things that are stupid but still work. It happens! But there should be some consciousness of this – at least there should if this is an expert who normally would not make such mistakes. Omar might, say, have displayed nervousness, rather than casually smoking a cigarette and being witty with his victims, as if he were just an actor and already had read to the end of the scene and didn’t have to worry since he knew he’d prevail. But that’s Hollywood; and in this department that’s what they were imitating.
Also, a thing about top gunfighters is that they can actually hit their targets, even from a serious distance. There’s a scene where Omar has a clean shot at Avon Barksdale, but instead of just shooting him, he walks close and takes so much time in doing so that he gets interrupted. Okay, a fifty-yard shot (that being something like the original distance) isn’t entirely trivial with a pistol, but a good pistol shooter can hit a stationary man-sized target at that distance. Or Omar could have brought a rifle, with which such a shot is trivial, or even just loaded rifled slugs into his usual shotgun. (The entire concept of rifles is foreign to The Wire – mostly for good reason since pistols are the usual weapon of inner-city violence, being preferred for their concealability. But guns are much easier to hold on target when you can put them to your shoulder, which is why rifles are the thing for long shots. A top-level gunfighter going after a heavily-guarded drug kingpin wouldn’t do it any other way.)
The question of accuracy also comes up in the first confrontation between Mouzone and Omar, where Mouzone is talking about hitting Omar in the elbow from what looks to be about 15 yards. Now hitting a fast-moving elbow from that distance really is somewhere between very difficult and impossible, especially if shooting from the hip (his stance at the time). It’s a lot more difficult than hitting a near-stationary torso at 50 yards. Of course it’s Mouzone saying that, not Omar, and it’s just boasting rather than demonstrated skill; but the scene doesn’t seem intended to show the two as stupid boasters, even though what they are saying is absurd. (Omar says in return that with his .45 “Even if I miss I can’t miss”.)
As regards having an army to back him, in his first action Mouzone does have one – he’s being called in to help the Barksdale “muscle” – but unaccountably doesn’t use it. Instead of coordinating with them, he just marches up to a housing tower and runs off a competing gang, all on his lonesome. (Well, with a sidekick, but one who is a personal servant, and a mediocre one at that, rather than someone skilled at arms.) Then he sits outside reading, supposedly making sure they stay away – when really they could have just shot him from a window, or for that matter had someone walk up behind him and blast him. There is a reason gangs are gangs and armies are armies: teamwork is important in fighting, as Patton most memorably put it</a>.
Anyway, this unreality is confined to those two supposedly top-level shooters; when it comes to the everyday gunplay of the drug trade the show hits the nail straight on the head. The shooting with the guns held sideways; the spray-and-pray that often leaves bystanders injured or killed – all that they do a great job on, as can be expected from their deep experience in the Baltimore homicide scene. But they’re not gun people, so turned to Hollywood when it came to the two top-level shooters. Of course even with those two characters, in many instances they just used their common sense, which yielded much better results. Indeed, in the third season they introduced Marlo’s shooters, who are not modeled on Hollywood and are much more realistic. They actually shoot at practice targets, for instance, rather than just relying on their macho studliness to keep their motor skills sharp.
My other criticism is not in the same league, since I don’t think the show’s creators could have done any differently. But there’s a problem with their street characters: to use Marc MacYoung’s language, “they don’t have the stink”. As he explains</a>:

There is a certain psychic odor that comes from growing up and living in the streets. It’s a rot that comes from constant exposure to violence, death, alcoholism, drug addiction, sociopathic behavior, poverty, sadism and viciousness. It’s reflected in a person’s attitudes, speech patterns, personal interactions and how he looks at the world. It’s a certain hardening of the spirit that comes from living years with the attitude of “do unto others before they do unto you”; Add to that the chronic paranoia of having spent years looking over your shoulder, lest vengeance-minded people you have wronged, slither out of that shadow you just passed. </blockquote>
Ordinary actors don’t have “the stink”, however stinky the language you put in their mouths; and the show used mostly ordinary actors from ordinary middle-class backgrounds. Being from such a background myself, I might not have noticed, except that they actually did use some street people as actors, and the contrast is stark.
One of them is “Snoop”. With the other female characters, it’s instantly apparent that they’re female, but Snoop had me uncertain for quite a while. This isn’t a transgender thing; the show was filmed before the transgender craze, or at least before it went mainstream. It’s a street thing: a hardness of spirit. It’s someone who has lived her life where to be female is to be a weak victim, and who hides her femininity in all sorts of little ways. Her manner is chilling: when she talks of violence and death, it’s with a flat matter-of-factness rather than with the ordinary displays of emotion. It renders her character scary in a way that many of the other characters are supposed to be but don’t quite pull off.
Another real street person plays Mouzone’s sidekick, and with an equally flat affect – though since he doesn’t engage in violence or even talk much, the result isn’t scary so much as wooden.
If all the characters that should be like that were like that, it would render the show somewhat repulsive. Their mannerisms alone would make ordinary viewers leery of them; viewers might even end up sympathizing with the cops who beat them up at every opportunity. That’s not the effect David Simon was trying for: he wants us to look past superficial mannerisms and really see how the street characters are victims of their circumstances… and what easier way than normalizing those mannerisms, especially when normal actors do that by default?
That intention is somewhat debatable. It is good that viewers see that there are hard constraints driving the street, not just people doing bad things because they’re bad people and like to do bad things. And in an immediate sense it’s often true that they’re victims of their circumstances: when you’re a drug dealer you sometimes just have to shoot someone. But it’s not like people are unaware what a violent world they’re getting into when they become drug dealers.
Also, the street mindset is more than just superficial. Indeed, it would be hard to put on a show with lots of actors from the street: they aren’t reliable enough. Filming a movie or a TV show like this is a high-pressure event, where if an actor doesn’t show up he’ll be wasting tens of thousands of dollars of other people’s time; likewise if he shows up drunk or stoned. (Even in scenes where someone is supposed to be drunk, they normally have him just mimic being drunk.) Obviously it worked to pick out a couple of the best of them to put on the show, but having half the cast be flaky would be unworkable.
And a show has to be a show: there’s no point in making one that nobody wants to watch. Something truly repulsively realistic likely wouldn’t repay the money that went into making it – though it would be mighty curious to see what such a transformation would do to the aristocracy of the drug business, like Stringer Bell: how far do they need to display the emotional scars of street life? I’m really not sure, but the way they casually order people killed would be more plausible when those people totally looked and acted like they were from “a zip code that does not fucking matter”, as opposed to that just being a line delivered by a cynical cop.
So this second criticism is really not so much a criticism as an observation: they did a lot of good work to make the show realistic, but there were still limits.

This inflation

Tue, 17 Sep 2024 00:00:00 +0000

The cause of the recent inflation seems clear, but a lot of people are getting it wrong. Kamala Harris is blaming price gouging, dismaying even people who are trying to cheer for her. Donald Trump, in his interview with Elon Musk, pointed to oil price increases as the main driver of inflation – which is plausible enough to deserve an actual refutation, since transportation prices do drive up the price of other things, and oil is also a chemical feedstock. But the early-2000s oil price jump was much larger, with the price of oil roughly tripling, yet the effect on overall inflation was much less than the 20% of the Covid years. 
Elon blamed government spending, which is a lot closer to the truth; but still, what directly drives inflation is printing money. Government spending doesn’t have to be done by printing money: it can be done by taxing or by borrowing, neither of which is particularly inflationary: they have major downsides, but inflation is not one of them. But when the government prints money, the result is direct: the more money there is, the less each bit of it is worth. (“Printing money” these days is mainly done electronically rather than with ink and paper, but the result is the same.)
In this case, the government did indeed finance the Covid lockdowns by printing money. They didn’t really have much choice: one of the first things that happened when it became clear Covid would be a serious problem was a market crash as speculators anticipated economic troubles. (“Speculators” is a dirty word in leftist circles, but thinking of the future is an inherently speculative activity, so I use the word here in a neutral sense.) The crash was stopped by the Federal Reserve, essentially by printing money and using it to buy government bonds. This did not produce immediate inflation: the private sector was decreasing the quantity of money – financial crashes do that – so the newly printed money just compensated for it. But the fact that this intervention was necessary made it clear that financing the Covid lockdowns by borrowing wouldn’t have worked: it would have meant selling government bonds, the reverse of that action, likely provoking another crash.
It is entirely possible and reasonable to wish for an economy that is not so fragile – not so highly leveraged, and thus not so needy of Federal intervention involving printing more money. But when the government finds itself with a highly-leveraged economy and a market crash, it has to intervene; the only alternative is deleveraging via the crash, which is hugely destructive even of things that do not at all deserve destruction. Hard decisions have to be made as to which investments to give up on, and in a financial crash there’s just no time to make those decisions properly.
To pay for the lockdowns, the government disbursed trillions of dollars, borrowing the money, but with the Federal Reserve printing money to buy government bonds: a two-step process, but one that resulted in about the same outcome as if they had just printed the money and spent it.
The upshot is that to argue against inflation one really has to argue against the lockdowns. One could try to argue against lockdown subsidies rather than lockdowns themselves, but that would mean insisting that people should have been prevented from working and yet not provided with any compensation: that they should have just suffered. Of course many could have lived off savings, and many more could have borrowed. But sudden mass borrowing by individuals has similar effects to sudden mass borrowing by the government; and sudden mass withdrawal of savings does something very similar, since savings are what banks lend out. So even those would have led either to a crash or to more money being printed. The only real way to have avoided that would be the way people with no savings and no credit would be forced into: suffering. That’s a relative term: a lot of what Americans consider suffering is just normal life for many in foreign countries – so to call it suffering is not to absolutely rule it out; it’s just to say it would be political suicide for any politician today to advocate for it.
During the lockdowns themselves, prices were also driven up by actual shortages, the result of industries being shut down. But that does not explain prices remaining high even now that those shortages are past; that’s due to the printing of money.
The lockdowns were initially sold as “just two weeks to flatten the curve”, but that was never good math: there was no reason to think that after two weeks were up the situation would be much different. The number of cases would be reduced, but with containment relaxed, exponential growth would quickly restore it. So after two weeks, they looked again at the situation, and the lockdowns weren’t lifted… but without apologizing for the initial “just two weeks” sales pitch, supplying a new rationale, or even saying how much longer it would be.
Frankly, what we were doing with the lockdowns was aping China. Western epidemiologists had had a doctrine that respiratory diseases were unstoppable and so it wasn’t worth trying (despite the first SARS having been eradicated globally). But lockdowns had worked for the Chinese, so we imitated them. With manufacturing, China has often been accused of copying Western ideas – largely fairly, but also largely unfairly, since turning ideas into practice is usually the hard part. (They often did also copy the turning-into-practice part of our knowledge, and even had considerable Western help, which is why the criticism as a whole has a lot of fairness to it.) But here the shoe was on the other foot: we copied them, and so poorly that copying isn’t even the right word, since we didn’t copy the key ingredient: the killer instinct.
A lot of things can be muddled through, making constant compromises between two competing views, dithering, going back and forth, adopting ideas from one side or the other. But with pathogens, either you’re trying to eradicate them or you’re not. The Chinese were trying for eradication, and succeeded. When mild measures didn’t work, they ramped up the intensity. In Wuhan in the early days, disinfection trucks roamed their streets emitting clouds of disinfectant – probably not usefully, but it was something to try, and was in the right direction, so they tried it. Their propaganda had them “building hospitals” in ten days, which was always a bit ridiculous: yes, the Chinese build fast, but they’re not superhuman; real hospitals, fully laid out in hospital fashion and filled with real hospital equipment, take more than ten days. What they really built were better described as warehouses for mild cases: people who didn’t need any serious health care but were being forcibly isolated so that they couldn’t spread the disease.
Throughout the pandemic, we chose the weaker form of isolation: telling people who tested positive to stay at home. The Chinese tried this at first, very quickly found that it didn’t work well enough, so shifted to forcible confinement. At some times and places, even people who weren’t sick were not allowed to go out on the streets. There were videos of lone walkers being admonished by drones with loudspeakers for being outdoors. Apartment buildings had doors and gates welded shut so that people could not get out. (Presumably they left one entrance open but guarded: still a severe hazard from a fire safety point of view, but worth it if done only for weeks to eradicate a killer disease.) They tracked people via their cell phones, forcing everyone to run an app which assigned each person a color-coded risk status based on where they’d been.
In the West there was an Apple/Google initiative to do cell phone contact tracking, in anonymized fashion, via Bluetooth sensing of nearby phones. It was something that people were suspicious of, and didn’t get used much. Suspicion was indeed appropriate, but here was misdirected: the publicized initiative wasn’t actually the problem. Indeed, it erred on the wimpy side. The information it delivered wasn’t very useful: it told an app user if another app user who had been near them had just been diagnosed with Covid, relying on users themselves to tell the app that they’d just been diagnosed. It could only really have been useful if everyone had the app and was conscientious about telling it about their diagnosis. Even then, “you might have been infected” is not a particularly useful piece of information for the victim to know; and (due to privacy concerns) the victim was the only person notified. The Chinese app, in contrast, was made mandatory and was required as a sort of passport for travel.
The reason suspicion of Apple and Google was actually appropriate was that both companies were already tracking their users and could easily have parlayed that tracking data into a Chinese-like risk assessment / method of control. This anonymous app initiative was just a fig leaf to hide their vast tracking databases, to preserve those for their really important use: selling advertising.
Throughout, the Chinese realized that eradication was the only worthwhile goal. This applied even at the end: when Omicron defeated their control measures, outside observers were astounded that rather than abandoning those measures gradually like we had done, they did so completely and all at once, letting the disease run wild. But it was the right move: those measures could only be justified as part of a complete package aimed at eradication.
The Chinese did wait far too long to abandon control: they battled Omicron for months with harsh lockdowns in cities that were vital to their economy. But before Omicron, for more than a year, they achieved nearly full eradication, letting life go on normally. There is Chinese propaganda that Covid was originally introduced from the West via frozen food; this is not at all plausible as regards its original origin, but entirely plausible as regards the occasional outbreaks they had during the period where they’d otherwise eradicated the disease. Transmission via frozen food is too rare to see in our own statistics; but when the disease is eradicated locally, even rare forms of transmission from foreign countries become evident.
Besides Omicron’s increased transmissibility, there was another reason for it to signal relaxation of control measures: it is much less deadly, inflicting about a quarter the death toll of the original variant, putting it in flu territory. One of the great failings of public health information has been not telling people this, with the result that many people still think Covid is a bigger menace than it currently is, and many others think that it was never a big menace.
This mortality difference is no secret; there are published papers (such as this one</a> or this one</a>) on it. Those papers do control for prior immunity: though Omicron is good at infecting people with prior immunity (whether from vaccines or from infection), and although that is a big reason why the Omicron wave infected so many more people yet was so much less deadly, it’s not the explanation for the results of those studies; they tease out the inherent deadliness of the strain in people of equal immune status.
Even the mechanism for the difference is known: Omicron enters cells in a different way (not using the enzyme TMPRSS2), giving it more affinity for the upper respiratory tract and less affinity for the lungs. Becoming more of an upper respiratory tract disease explains the greater transmissibility: the nose is an easier place for germs to get to (and an easier place for them to leave) than the lungs are. But lung infections are more deadly.
Though no secret, this still represents a failure in public messaging: those responsible (both in government and in the press) were too scared to share the good news. I heard it in good time from one of Peter Attia’s podcasts</a>, but those are technical discussions which don’t reach a mass audience.
So though the Chinese waited too long to end their control measures, they probably saved a lot of deaths from the pre-Omicron variants. But our “lockdowns” were much less severe than the Chinese ones, so they didn’t come anywhere near eradicating the disease. Large numbers of “essential” businesses were allowed to keep working, including liquor stores. Whatever impetus there was towards making restrictions more severe was lost in the George Floyd riots: conservatives looked at the tolerance extended to rioters’ lockdown violations and concluded that the people pushing for lockdowns weren’t really serious. Even from a nonpartisan point of view the riots served as a reminder that we have a large and unruly lower class, accustomed to lawbreaking, which would serve as a reservoir of infection even if the upper and middle classes did everything right; it would have taken something like martial law to change this.
People like Anthony Fauci are commonly blamed for the lockdowns, with Elon Musk even calling for him to be prosecuted. But being a public health jackass is not actually a crime. It wasn’t even Fauci’s job to consider the economic downsides of lockdowns; that’s the job of politicians. They needed to take the attitude that Clemenceau exhibited during World War 1 when he said “War is too important to be left to the generals”. Public health authorities aren’t experts on the economy, so can’t be trusted to weigh economic factors against health factors; that’s their bosses’ job.
The lockdowns continued for quite a while. No definite cutoff can be named, since they were gradually relaxed and the schedule varied from state to state. But six months is something like a ballpark figure for the half-life of the whole lockdown phenomenon in the US. This had to be paid for: the first federal spending bill allocated two trillion dollars, and subsequent bills added more trillions to that. Compared to the initial talk of “two weeks”, six months cost quite a lot.
In this, both parties have a large degree of complicity: at first both fully signed on to it. Republicans were first to start murmuring, and then all-out opposed the final dose of pork, but perhaps only Thomas Massie can boast (as he does 46 minutes in to this interview</a> with Tucker Carlson) of having opposed lockdown subsidies from day one. All this makes the inflation an ideal political football: neither party can truly get to the root cause since that would implicate them too. Also, by now the inflation is pretty much over, killed by the Fed’s raising of interest rates, so whatever nonsense politicians now promulgate as a remedy, they’ll be able to announce victory.

Lawyers' privileges

Wed, 28 Aug 2024 00:00:00 +0000

A recent post</a> by Eugene Volokh describes a legal case which hinges on a rule of libel law that most people are likely unaware of:

“It has long been the law of Pennsylvania that statements made by judges, attorneys, witnesses and parties in the course of or pertinent to any stage of judicial proceedings are absolutely privileged and, therefore, cannot form the basis for liability for defamation.” </blockquote> 
Volokh makes no comment of his own regarding the merits of this rule, but I have seen a devastating comment on rules of that sort, made by Thomas Babington Macaulay in his Notes on the Indian Penal Code. The setting is not quite the same, but it’s close enough:

… the crime of deliberately and knowingly asserting falsehoods in pleading. Our opinions on this subject may startle persons accustomed to that boundless license which the English law allows to mendacity in suitors. On what principle that license is allowed we must confess ourselves unable to discover. A. lends Z. money; Z repays it. A. brings an action against Z. for the money, and affirms in his declaration that he lent the money, and has never been repaid. On the trial A.’s receipt is produced. It is not doubted. A. himself cannot deny that he asserted a falsehood in his declaration. Ought A. to enjoy impunity? Again: Z. brings an action against A. for a debt which is really due. A.’s plea is a positive averment that he owes Z. nothing. The case comes to trial; and it is proved by overwhelming evidence that the debt is a just debt. A. does not even attempt a defence. Ought A. in this case to enjoy impunity? If, in either of the cases which we have stated, A. were to suborn witnesses to support the lie which he has put on the pleadings, every one of these witnesses, as well as A. himself, would be liable to severe punishment. But false evidence in the vast majority of cases springs out of false pleading, and would be almost entirely banished from the Courts if false pleading could be prevented.
It appears to us that all the marks which indicate that an act is a proper subject for legal punishment meet in the act of false pleading. That false pleading always does some harm is plain. Even when it is not followed up by false evidence it always delays justice. That false pleading produces any compensating good to atone for this harm has never, as far as we know, been even alleged. That false pleading will be more common if it is unpunished than if it is punished appears as certain as that rape, theft, embezzlement, would, if unpunished, be more common than they now are. It is evident also that there will be no more difficulty in trying a charge of false pleading than in trying a charge of false evidence. The fact that a statement has been made in pleading will generally be more clearly proved than the fact that a statement has been made in evidence. The falsehood of a statement made in pleading will be proved in exactly the same manner in which the falsehood of a statement made in evidence is proved. Whether the accused person knew that he was pleading falsely, the Courts will determine on the same evidence on which they now determine whether a witness knew that he was giving false testimony.
We have as yet spoken only of the direct injury produced to honest litigants by false pleading. But this injury appears to us to be only a part, and perhaps not the greatest part, of the evil engendered by the practice. If there be any place where truth ought to be held in peculiar honour, from which falsehoods ought to be driven with peculiar severity, in which exaggerations, which elsewhere would be applauded as the innocent sport of the fancy, or pardoned as the natural effect of excited passion, ought to be discouraged, that place is a Court of Justice. We object, therefore, to the use of legal fictions, even when the meaning of those fictions is generally understood, and we have done our best to exclude them from this code. But that a person should come before a Court, should tell that Court premeditated and circumstantial lies for the purpose of preventing or postponing the settlement of a just demand, and that by doing so he should incur no punishment whatever, seems to us to be a state of things to which nothing but habit could reconcile wise and virtuous men. Public opinion is vitiated by the vicious state of the law. Men who, in any other circumstances, would shrink from falsehood, have no scruple about setting up false pleas against just demands. There is one place, and only one, where deliberate untruths, told with the intent to injure, are not considered discreditable, and that place is a Court of Justice. Thus the authority of the tribunals operates to lower the standard of morality, and to diminish the exteem in which veracity is held; and the very place which ought to be kept sacred from misrepresentations, such as would elsewhere be venial, becomes the only place where it is considered as idle scrupulosity to shrink from deliberate falsehood. </blockquote>

Reparations

Mon, 12 Aug 2024 00:00:00 +0000

Reparations for slavery were paid a long time ago.
They were paid in blood, perhaps the only suitable currency for such a payment. As Lincoln put it, “…every drop of blood drawn with the lash shall be paid by another drawn with the sword…”. 
In context, that was an unwanted possibility rather than a threat or a promise; but by the time he spoke those words in his Second Inaugural speech it likely had already happened. I’m not going to attempt a calculation of drops of blood (or gallons or metric tons of it) – I’ll leave that to those who are braver about making up numbers – but the Civil War was the most lethal of any war this country has fought, counted in absolute numbers of people killed. Since the population has increased greatly since then, in percentage killed it towers even higher over subsequent wars. And it was white men who bore the brunt of the fighting.
Everything that is talked about as possible “reparations” these days is puny in comparison.

The characters of plastics

Fri, 09 Aug 2024 00:00:00 +0000

“It’s made of plastic” is a common put-down. Marketers selling higher-end bits made of plastic (like gun parts) try to evade the stigma by calling it “polymer”, but that’s just a stupid euphemism: every plastic is a polymer, though not every polymer is a plastic. The word “polymer” says nothing to indicate that this might be a superior sort of plastic. Yet there are superior sorts; plastics vary widely in their characters. Some analogies between plastic and human characters: 
Polyethylene, polypropylene: Snow White (simple, pure, weak). These are the cheapest and most common plastics, and are made of just hydrogen and carbon atoms. Ethylene (two carbons and four hydrogens) is polymerized to make polyethylene, and the quantity of ethylene made each year is measured in cubic miles. They can of course have pigments added to them to give them color, but are commonly used in uncolored, nearly-pure form. Chemically they are unreactive, which makes them good for containers of all sorts; they are even used for chemistry beakers. That same unreactivity makes them very hard to glue, and means they don’t deteriorate with time. Leave them out in sunlight, though, and the UV quickly weakens them to where they crack easily.
Nylon: Arnold Schwartzenegger (strongman). Plastic nuts and bolts, which need to be strong, are usually nylon, as is monofilament fishing line and womens’ hosiery (delicate, but still strong for its weight). Nylon adds nitrogen to the list of atoms it contains; it’s chemically known as a “polyamide”, a class which also includes proteins.
Nylon with 30% glass fiber reinforcement: Arnold Schwartzenegger on steroids. It’s as strong as cast aluminum (though extruded aluminum can be much stronger). When guns are made of plastic, this is usually the plastic they use; likewise for the plastic housings of electric drills and other power tools.
Polycarbonate: Achilles (warrior with a fatal weakness). Polycarbonate is what they make bulletproof windows from, and safety glasses, and the transparent fronts of car headlights. Its weakness is chemical attack: a splash of acetone, and it instantly “crazes”, a network of fine cracks appearing across its surface as built-in stresses are relieved. (If the part has any built-in stresses, that is; molded parts probably do, but flat sheets might not.) Polycarbonate is often protected by a surface coat to block chemical attacks. It’s the main plastic that gives off the notorious bisphenol A (BPA), which is probably not so bad as it’s reputed to be. But it still doesn’t make sense for food containers to be made from polycarbonate, since they don’t need to be bulletproof and polycarbonate is pricey.
Polyvinyl chloride (PVC): Dr. Jekyll and Mr. Hyde (a character who varies wildly depending on which drugs he takes, with more than a bit of evil in him). PVC is normally rigid, and is used in that form for house siding and sewage pipes, but can be pumped full of plasticizer to make it flexible; in that state it’s used for inflatable boats, shower curtains, imitation leather, and even sex toys. In pure form it’s not a stable chemical; its staying good depends on added “stabilizers”, which themselves can be somewhat evil: they often contain the toxic element lead, which normally is locked in the plastic but is unleashed when the plastic deteriorates or is burned. Burning it also gives off a variety of toxic chlorine-containing substances. The plasticizer, if used, evaporates away over years (in cars, landing on the inside of the windshield and producing an annoying haze which has to be wiped away), and eventually the remaining plastic gets brittle and cracks. The plasticizer is often a phthalate, another notorious class of chemical and also probably not as bad as popular repute would have it, but still not good for you and a pain to clean up. (People commonly regard soft plastics in cars as good and hard ones as bad and cheap, but I don’t think they’ve made the connection between having soft plastics around and needing to clean off the inside of the windshield, or for that matter having phthalates in the air they breathe when getting into a car on a hot day.)
Polyurethane: an android from the movie Blade Runner (strong, versatile, dangerous, doomed). Polyurethane can be made either hard or flexible, with the flexible varieties used to make rubbers and foams, but polyurethanes have a tendency to decide “now it’s time to die</a>”, losing almost all of their strength, with foams crumbling or turning to goo and rubber parts cracking. They are made using isocyanates, in a reaction which could be thought of as Nazi chemistry since it was in fact invented in Germany during that era and since isocyanates are quite toxic. (That’s just a resemblance, of course; in reality Nazi ideology had nothing to do with chemistry.) Though toxic (methyl isocyanate killed thousands of people in the Bhopal disaster), isocyanates are not cyanides (which are even worse), and in finished polyurethane there are only traces of isocyanates left, though they might be the cause of the annoying smell that new polyurethane foam often has. But burning polyurethane does produce some hydrogen cyanide, and polyurethane foam burns unusually vigorously, unless it’s been treated with flame retardants, which thus have been mandated by law in some places but which themselves might be a health issue.
Bakelite, aka phenol-formaldehyde: well, nobody really comes to mind as a corresponding character, but it’s rigid, brittle, and can take a lot of heat. Bakelite was one of the first plastics, and is thermosetting: it doesn’t melt but rather has to be formed from its ingredients (phenol and formaldehyde) into its finished shape (requiring a mold, heat, and pressure). With it we’re back to things that contain only the safer elements (carbon, hydrogen, and in this case oxygen). Phenol and formaldeyhde are each nasty, but their nastiness is consumed when they react together. Bakelite’s brittleness can be mitigated by using appropriate fillers, particularly fibrous ones. Its heat resistance is such that, mixed with high-temperature fibers, it is used for heat shields for reentry from space (which burn away but do so slowly enough that they don’t burn through). It also sees a lot of use in things like electrical circuit breakers, where other plastics might melt into a blob and let their metal parts short-circuit.
Polystyrene: Joe Sixpack (common, cheap, weak, nasty). It’s not heat-resistant, not chemical-resistant (it dissolves in a wide variety of solvents), and is brittle. But it can easily be blown into a foam (styrofoam), which because of the bubbles is a good insulator; being a foam also mitigates its brittleness. When heated it de-polymerizes and gives off styrene, which has a distinctive acrid smell. It can be improved into “ABS” by adding large proportions of acrylonitrile and butadiene into the polymerization process; ABS has the same distinctive smell when heated but is considerably tougher, making it suitable for a wide variety of consumer products (though it’s still much weaker than nylon). Legos are ABS. If there’s a plastic which really deserves the putdown “it’s made of plastic”, it’s ABS: it’s common enough to be the sort of thing people think of when they hear “plastic”, and is generally mediocre. But as human analogies go it’s a step above Joe Sixpack; call it Joe Blow.
Teflon: Queen Victoria. “Nobility” in chemistry means a lack of reactivity, an immunity to chemical attack; “noble metals” are noble not because they’re expensive, but because they snootily look down their noses at other chemicals and refuse to have relationships with them. In plastics it’s harder to get nobler than teflon (polytetrafluoroethylene; PTFE), which differs from polyethylene by having all its hydrogens replaced with fluorines, which are much more difficult to dislodge. Fluorine chemistry being a difficult and dangerous sort of chemistry, PTFE doesn’t come cheap. It’s used to coat nonstick pans, where its nonreactivity translates into things not sticking to it. As best I can tell (though information on this is scarce), even the newer “ceramic” nonstick coatings often have an imperceptibly-thin layer of a molecule with a highly-fluorinated tail that resembles PTFE; the ceramic gets the publicity but the highly-fluorinated chemical does the work of making the coating nonstick… until it wears off, which because of the thinness of the layer happens more quickly than it does with the older sort of pan which has a visible layer of PTFE. (Modern nonstick pans are often regarded with suspicion; but the old mainstay in that department, cast iron, gets its nonstickness by being “seasoned”: coated with a layer of burnt, polymerized oil, no doubt containing many carcinogens.)
Those are just some highlights of a complicated subject; I haven’t even mentioned some major plastics, and there are a host of minor ones, as well as innumerable subvarieties of the major ones. (Take polyethylene, normally weak, react it until the molecular chains are long, then stretch it until they are aligned, and you get an extremely strong fiber, suitable for high-strength ropes or bulletproof vests: not Snow White but Wonder Woman.) With such a varied cast of characters, it’s a pity that they all get lumped together as “plastic” in common usage. I suppose it’s somewhat inevitable, since you can’t just look at a piece of plastic and tell what sort it is. But more public awareness of the differences would be nice.

The Bull

Tue, 06 Feb 2024 00:00:00 +0000

The argument is often made that although some group committed a violent aggression, they were provoked into it by an act that was like waving a red flag in front of a bull. The South at Fort Sumter; the Russians in 2022; the Nazis: all those have had that argument made for them. 
The people who make that argument too often forget that the red flag only works because the bull is a bull. Humans react to the waving of a red flag by asking “Why are you waving that thing at me?” There are humans who act like bulls, but one does not get to turn around and proclaim that they are “playing four-dimensional chess”, as some people liked to say about Putin. Bulls don’t do that; they trample chessboards and shit on them. (And, indeed, the “four-dimensional chess” line seems to be taking a bit of a rest at the moment.)
That doesn’t mean the argument has to be wrong. A historical case where it is generally accepted is the Ems telegram, which provoked the French to declare war against Prussia in 1870. But that general acceptance doesn’t come with the idea that the French were actually right to declare war, or even that they were “in the right” though they lost the war. They were just being foolish.
So the argument can be right, but it can’t do what most of the people making it want it to do, which is to reverse the morality of the situation. To make it, you have to start by assuming the aggressor is mentally inferior. Then, well – yes, bulls have their uses and don’t always need to be fought.